INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 carrier

-0.79

iverpool

-0.76

fal

-0.65

 carriers

-0.65

usable

-0.64

ounge

-0.63

borough

-0.63

omorph

-0.62

estation

-0.62

asus

-0.62

POSITIVE LOGITS

 Cosponsors

0.80

Ã©n

0.78

oshenko

0.75

hiro

0.73

 Refuge

0.73

estern

0.70

 Scores

0.67

 Transcript

0.65

 Fifth

0.64

Newsletter

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.